Performance of six influenza rapid tests in detecting human influenza in clinical specimens

https://doi.org/10.1016/j.jcv.2007.03.002Get rights and content

Abstract

Background

The rapid diagnosis of influenza can alter the management of a patient's illness, resulting in reduced antibiotic usage, correct use of influenza antivirals and reduced length of stay in hospital emergency departments. The rapid tests have also been used to detect outbreaks in institutions and may play a role in pandemic influenza control.

Objectives

To test six different rapid influenza tests, in a head-to-head comparison for the detection of seasonal influenza types A and B, compared to laboratory-based tests.

Study design

One hundred and seventy-seven clinical specimens taken from mostly paediatric patients between June and October 2006 were tested using six influenza diagnostic tests and three laboratory-based techniques (immunofluorescence, cell culture and real-time RT-PCR).

Results and conclusion

Compared with cell culture, five of the rapid tests (Binax Now Influenza A&B, Directigen EZ Flu A + B, Denka Seiken Quick Ex-Flu, Fujirebio Espline Influenza A&B-N, and Quidel QuickVue Influenza A + B Test) demonstrated a similar influenza A sensitivity of between 67–71% and a specificity of 99–100%, however one rapid test (Rockeby Influenza A Antigen Test) had a significantly lower influenza A sensitivity of only 10% (specificity was 100%). For the five kits that detected influenza B antigen, sensitivity was considerably lower than that seen for influenza A (sensitivity for all the kits was 30%), although the number of specimens containing influenza B viruses was low.

Introduction

Influenza is a respiratory infection that can result in significant morbidity and mortality, particularly in the young, elderly or immunosuppressed (Simonsen, 1999). However, rapid diagnosis of influenza infection can facilitate the use of appropriate treatment, both improving the patients clinical outcome and significantly reducing hospital costs (Bonner et al., 2003). A large number of rapid influenza detection tests are now available that are simple to use, quick (generate a result in 15 min) and can be performed outside the laboratory. In general, previous studies have reported the tests to have a good specificity (99–100%), however different studies have reported significantly different sensitivity values even when evaluating the same test (Storch, 2003). For example the reported influenza A sensitivity for the Becton Dickinson Directigen Flu A + B Test ranges from 44% (Cazacu et al., 2004) to 96% (Chan et al., 2002). While study-to-study variability in sensitivity with the same test can be explained by the differences in specimen type, influenza type/subtype circulating, patient age and the ‘gold standard’ test used for comparison, these differences highlight the difficulties in comparing results of individual rapid tests when they are evaluated in different studies. To date there are few reports of evaluations that have included more than three rapid tests within the same study (Rodriguez et al., 2002). In this study, however, we have conducted a head-to-head comparison of six influenza rapid tests for their ability to detect influenza A and B on the same clinical specimens and compared these results with laboratory detection methods for influenza.

Section snippets

Clinical specimens

One hundred and seventy-seven respiratory samples (150 nasopharyngeal aspirates; 8 nasal swabs; 8 bronchioalveolar lavages; 6 sputum samples; 5 throat swabs) were obtained from patients with influenza-like illness at the Royal Children's Hospital, Melbourne, Australia between June and October 2006. Patients were aged from 4 days to 64 years old (with 78% of patients being five years old or less), 59% of patients were male and 41% female.

Rapid influenza tests

Six influenza rapid tests—Binax Now Influenza A&B (Binax;

Results

Of the177 clinical specimens obtained, 49 were found to be positive for influenza A antigen by RETCIF; the laboratory based technique which has been used in this evaluation as the ‘gold standard’ for the purposes of rapid test comparisons. An additional specimen was determined to be influenza A positive by real-time RT-PCR (i.e. 50 influenza A positive specimens), while one less specimen was found to be influenza A positive by DIF (48 influenza A positive specimens). Only ten specimens in the

Discussion

Comparisons between the sensitivities of rapid tests evaluated in different studies are inherently difficult given the variability in study designs (patient age, sample type and comparators). As such, studies like this present one, which compared six different rapid influenza tests head-to-head on clinical samples, are of significant value to physicians, general practitioners and others in choosing which influenza rapid test to use. Apart from the Rockeby test, the other five tests demonstrated

Acknowledgements

The authors would like to thank the following for proving rapid test kits for evaluation in this study: Lab Diagnostics, Australia (Binax Now Influenza A&B); Becton Dickinson, Australia (BD Directigen EZ Flu A + B), Denka Seiken, Japan (Quick Ex-Flu), Fujirebio, Japan (Espline Influenza A&B-N), Rockeby, Singapore (Rockeby Influenza A Antigen Test) and Panbio, Australia (Quidel QuickVue Influenza A + B Test).

The WHO Collaborating Centre for Reference and Research on Influenza is supported by the

References (19)

There are more references available in the full text version of this article.

Cited by (0)

View full text